Newsgroups: comp.ai.philosophy,comp.ai,comp.robotics,comp.cog-eng,sci.cognitive,sci.psychology
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!gatech!rutgers!gw2.att.com!nntpa!not-for-mail
From: jcawkwel@mlsma.mlm.att.com (Jack Cawkwell)
Subject: Re: Grounding Representations: ("Grounding" is the wrong word)
Message-ID: <D8MCtF.523@nntpa.cb.att.com>
Sender: news@nntpa.cb.att.com (Netnews Administration)
Nntp-Posting-Host: mlw079.mlm.att.com
Organization: AT&T
References: <3nhlk5$i7o@percy.cs.bham.ac.uk> <3o072m$bes@jhunix1.hcf.jhu.edu> <799256142snz@longley.demon.co.uk> <799313168snz@chatham.demon.co.uk>
Date: Mon, 15 May 1995 12:17:39 GMT
Lines: 23
Xref: glinda.oz.cs.cmu.edu comp.ai.philosophy:28084 comp.ai:29858 comp.robotics:20725 comp.cog-eng:3150 sci.cognitive:7605 sci.psychology:41526


The conclusion to be drawn from the kittens brought up in a world of
vertical stripes, who where unable as adults to see horizontal stripes
surely supports the arguments against using the "grounding" concept.

The experiment demonstrates that perception is an active event, in which
the perceiver selects "knowledge" from the environment. This knowledge may
not exist before the act of perception. We see what we expect to
find, not what is there. We do not in general see what we expect to be
impossible. The history of science, invention and innovation is littered
with people who go beyond current paradigms and see what others previously
have not be able to.

This is a similar philosophical area to Kolmogorov complexity,
Goedels Incompleteness Theorems, and the Mind Body Problem. The best
sound bite for which is:
      recognition is about find the simplest representation.

-- 
Jack Cawkwell

AT&T Malmesbury SN16 9NA UK
Jack.Cawkwell@att.com +44 666 832295
